<p>Students vote with their feet. Even with the obvious cost differential, students having the choice, choose NU over ucb 53%/47%, Brown over ucb 86/14, Penn over ucb 77/23, Cornell over ucb 67/33, and, for example, MIT over ucb 95/5. Ucb beats Tufts, UVa and ucla. That should give you a good indication of where things stand. See pior post for source.</p>
<p>The NYT figure can be found here:
The</a> New York Times > Week in Review > Image > Collegiate Matchups: Predicting Student Choices
And the original paper can be found here:
SSRN-A</a> Revealed Preference Ranking of U.S. Colleges and Universities by Christopher Avery, Mark Glickman, Caroline Hoxby, Andrew Metrick</p>
<p>Some figures seem to defy "common wisdom on the streets". For example, Cornell over Cal 67/33.</p>
<br>
<blockquote> <p>Even with the obvious cost differential, students having the choice, choose . . .</p> </blockquote>
<br>
<p>Since this is based on a nationwide survey and presumably not heavily weighted towards California, you are looking at OOS costs for Cal which means the cost differential is not a significant factor in these preferences.</p>
<p>These numbers are scewed because I doubt as many Californians would go to Ithaca and thus do not apply there in as great of numbers as East Coast folk do.</p>
<p>You can try to explain away the results, but there is no reason to believe that the nyt figures are "skewed" in any way. Why would you assume that people drawn to cornell apply in greater numbers at ucb than vice versa? Why would you infer that the ucb applicants reflected in the survey are OOS and paying the higher tuition? These are just rationalizations. I have an explanation..."State School". ucb is a good one and beats the other state schools on the list ucla and UVa. Mich and UNC werent charted. But for undergrad it just doesnt stack up to the great (or even very good) private schools. This is a fairly efficient market in that there is alot of relevant information concerning all of these schoools out there and therefore actual student choices are significant. Admissions offices track these statistics and, in many cases, try to devise strategies to win the competion for these students having multiple admissions. It looks like ucb has some work to do.</p>
<br>
<blockquote> <p>Why would you infer that the ucb applicants reflected in the survey are OOS and paying the higher tuition?</p> </blockquote>
<br>
<p>That is easy. This was a nationwide survey of seniors. Any sane survey would ensure that the results weren't skewed with a disproportionate bias to California. Ergo, majority of the respondents were not from California.</p>
<p>If a majority of the respondents were not from California, then the data would tend to skew the data AGAINST UCB unfairly because it's looking at cross-admit data from a pool of applicants that are a minority to a school like UCB. It would make more sense to look at a cross-section that is more representative of the average applicant/acceptee pool to a university (i.e. more from the state of California). </p>
<p>If you were to look at the cross-admit battles between acceptees from within California, I bet you would see a higher "win" percentage of UCB. If a majority of the applicants in this survey are from outside California (which comprises a minority in the pool of acceptees to UCB), then the cost benefits of a public university are diminished if not altogether eliminated. This is relevant!!! Universities provide many benefits, costs being one of them. You cannot cherry pick this factor out when looking at cross-admit battles (there are many reasons a person chooses go to a university - you cannot arbitrarily remove one as a factor to look at). </p>
<p>Therefore, in these nationwide surveys, private universities are likely to win over Berkeley because the private universities may cost similar to or even less than Berkeley (the assumption being that the private university can provide more benefits than the public university can for the same cost). </p>
<p>I believe that public universities are disadvantaged in cross-admit battles in nation-wide surveys because they don't accurately represent the majority of the pool of applicants to those public universities. Because public universities tend to favor in-state students AND charge a lot more to OOS students, they will tend to attract a lot more in-state students. That's the nature of the public university beast though I guess - whether for good or bad. Private universities on the other hand don't provide preference to in-state students and charge the same whether a student is OOS or in-state. This means that the top private schools are likely to get a more geographically distributed sampling of applicants/acceptees from across the nation and internationally. Therefore, doing a nationwide survey with equal geographical distribution of respondents, the results would likely be more accurate for the private universities when compared to other private universities. However, the data would be skewed against public universities in private vs. public university cross-admit battles for reasons mentioned above.</p>
<p>On the other hand, if you were to survey solely the in-state students of whatever state the public university is in, you'd likely see it as a cross-admit advantage. We obviously don't want to disproportionately skew it that direction either! </p>
<p>So in short, what I'm trying to say is, if a survey is being done to compare private and public university cross-admit battles, the respondents being surveyed need to better represent the actual applicant pool of those universities. For public universities, that means more in-state students should be surveyed. The number of OOS student surveys should not disproportionately be weighed.</p>
<p>This was reported as a single survey population that was then analyzed to report on the cross-school preference patterns. No ability in that kind of survey to choose separate populations appropriate to each type of school.</p>
<p>of these, i'd choose brown in a heartbeat</p>
<p>The New York Times article was based a study whose survey data were taken almost ten years ago (from December, 1999 to May, 2000). This seems a little out of date.</p>
<p>I agree with khan that a study like this can be totally skewed based on the population being surveyed. It would have been appropriate for them to list the geography, income levels, ethnicity of the ones surveyed for this study to make any sense.</p>
<p>"This was reported as a single survey population that was then analyzed to report on the cross-school preference patterns. No ability in that kind of survey to choose separate populations appropriate to each type of school."</p>
<p>I didn't say it would be easy to separate out the appropriate populations to survey for each type of school. My point still stands though - the results of a blanket nation-wide survey is not a very accurate meter for cross-admit battles between private and public universities. That survey unfairly makes the top private universities appear to win more cross-admit battles over public universities than actually occur. This is my point. </p>
<p>The survey population needs to more accurately represent the applicant pools of the respective universities. For public universities, a larger portion of the applicant pool will be in-state students.</p>
<p>
[quote]
You can try to explain away the results, but there is no reason to believe that the nyt figures are "skewed" in any way. Why would you assume that people drawn to cornell apply in greater numbers at ucb than vice versa? Why would you infer that the ucb applicants reflected in the survey are OOS and paying the higher tuition? These are just rationalizations. I have an explanation..."State School". ucb is a good one and beats the other state schools on the list ucla and UVa. Mich and UNC werent charted. But for undergrad it just doesnt stack up to the great (or even very good) private schools. This is a fairly efficient market in that there is alot of relevant information concerning all of these schoools out there and therefore actual student choices are significant. Admissions offices track these statistics and, in many cases, try to devise strategies to win the competion for these students having multiple admissions. It looks like ucb has some work to do.
[/quote]
</p>
<p>Actually, there are huge reasons to believe the results are skewed and even the authors of the report in the report itself refer to it as not being definitive or unskewed. It's funny that people who quite likely haven't read the report would be so adamant that it reflects no biases when even the authors allow for this possibility.</p>
<p>Here are some problems:</p>
<p>1) Look through the study and there are several different ways that they slice out their rankings and preferences. One is by region. So look at the region that includes Illinois, Michigan, etc. and where does University of Chicago rank in comparison to Berkeley, do you think? Intuitively, my guess would be that in California Berkeley would outrank Chicago and that in Illinois, Chicago would outrank Berkeley. And yet in the study, in the Illinois region, Berkeley outranks Chicago by a couple of places. In California, Berkeley only outranks Chicago by four places. No way. Cooper Union outranks them both -- in California -- by a considerable amount. Cooper Union? Who the f*** in California knows what Cooper Union is? Very few, I guarantee you. And yet Cooper Union outranks among others USC (!), Johns Hopkins, and Northwestern. Give me an effin' break. Any ranking of preferences that weights Cooper Union above USC in California has terribly serious bias flaws. I am not saying Cooper Union is bad. It's just that in California it's a non-entity. This is the same study that you are calling unskewed.</p>
<p>2) Another slice is to look at preferences in terms of intended fields of study. Again, let's look at Berkeley. Berkeley, MIT, and Stanford are the big three in terms of engineering, computer science, math and physics. Two years ago when science-focused Nobel prizes were handed out, one went to a Berkeley undergrad who went to MIT for grad school and who now teaches at Stanford, another went to an MIT grad who teaches at Berkeley, another went to a Berkeley grad who works at Nasa, etc.... Anyway, in this ranking, Cooper Union (again) and Colgate appear in a top-25 of a list that doesn't even rate Berkeley, a list focused on the "top choices of students who would pursue engineering, computer science, physical sciences or math." Again, give me an effin' break. That's even more absurd that the Cooper Union over USC in California ranking. Look around CC and see how accurate this. Colgate and Cooper Union, as wonderful as they are non-entities in a comparison against Berkeley in these fields. Anybody would see that.</p>
<p>3) The big problem with the study is essentially it goes to a bunch of people who'd be likely to choose certain schools, pools them, asks them what schools they'd be likely to choose, and surprise! it's just what one would expect with those kinds of a priori sample biases that others have alluded to.</p>
<p>The thing is the authors only mooted their ranking as the <strong><em>kind</em></strong> of ranking that could reveal more than some other rankings.</p>
<p>On the one hand there is a statistical study of actual student choices, done by a reputable organization that has no axe to grind. On ther other hand there are emotional criticisms and anecdotal comments of no provenance from partisans. Hmmmm, which to choose? </p>
<p>No doubt UCB is great for grad school and, along with Michigan and UVa, is among the great public universities. However, increasingly it lacks behind the great and near great privates in undergrad education. Among other things, there are too few resources; too many undergrads competing for them;Too much teaching by TAs; Too many large classes; No advising; Too much red tape; Getting closed out of classes; Poorer living conditions, etc. Students understand these things and no doubt they are reflected in actual student choices. As HYP etc continue to expand financial aid, you can expect even greater defection of top students to the privates.</p>
<p>Regarding your criticism concerning Cooper Union, you may want to consider that, notwithstanding financial resources that are far less than the Ivies or UCB, it is managing to offer free tuition to all applicants (likely an important factor in actual student choices) while building an entirely new physical plant. Contrast that to publicly funded state schools, which increasingly are raising tuition and nickel and dime students with other fees while limiting educational opportunities.</p>
<p>Hopefully, UCB will react positively to the changing market.</p>
<p>
[quote]
On the one hand there is a statistical study of actual student choices, done by a reputable organization that has no axe to grind. On ther other hand there are emotional criticisms and anecdotal comments of no provenance from partisans. Hmmmm, which to choose?
[/quote]
</p>
<p>You obviously didn't read the study. It absolutely had an axe to grind. Firstly, it was not based on actual student choices made to matriculate and the study itself criticized itself for not using actual matriculation data; it was based on a survey the sample for which was based on some questionable choices. It was trying to posit a new ranking methodology. But the study itself indicated that the rankings it created (and the ones quoted by the NYT) were meant to be an example. Not definitive. Again the authors of the study said so.</p>
<p>Here's the study:</p>
<p>
[quote]
No doubt UCB is great for grad school and, along with Michigan and UVa, is among the great public universities. However, increasingly it lacks behind the great and near great privates in undergrad education. Among other things, there are too few resources; too many undergrads competing for them;Too much teaching by TAs; Too many large classes; No advising; Too much red tape; Getting closed out of classes; Poorer living conditions, etc. Students understand these things and no doubt they are reflected in actual student choices. As HYP etc continue to expand financial aid, you can expect even greater defection of top students to the privates.
[/quote]
</p>
<p>Look who's got an axe to grind. I don't even care enough to show you the numbers indicating class sizes are most often usually not different between, for instance, Stanford and Berkeley. Just search for Kyledavid80 user name on here and look for some post. You're repeating things you've heard others say without checking the facts. Regarding resources, certainly several of the top privates have more resources on a per-student basis. Regarding access to quality academic resources, Berkeley's got it in spades.</p>
<p>
[quote]
Regarding your criticism concerning Cooper Union, you may want to consider that, notwithstanding financial resources that are far less than the Ivies or UCB, it is managing to offer free tuition to all applicants (likely an important factor in actual student choices) while building an entirely new physical plant. Contrast that to publicly funded state schools, which increasingly are raising tuition and nickel and dime students with other fees while limiting educational opportunities.
[/quote]
</p>
<p>I think you didn't read what I wrote if you thought I was criticizing Cooper Union. What I was showing was a very strong built-in regional bias. Very few people in California have ever heard of Cooper Union. In a paper talking about "revealed preferences" there would be no revealed preferences or actual matriculation choices made toward Cooper Union matched up against many dozens of schools in California. And even more laughably true with regards to people who want to study math, computer science, engineering, or physics across the nation let alone in California is that they'd more likely choose Cooeper Union vs. Berkeley or even USC. It's not saying Cooper Union isn't great. It's saying the study has flaws.</p>
<p>cooper union? wth is that? sounds like a restaurant.</p>